Analytics UI is the experience layer of the data platform. It turns curated warehouse tables and governed metrics into screens that data analysts and business users can actually work with — iterating SQL in notebooks, composing dashboard tiles and publishing reports — without talking to backend services directly.
Responsibility in one sentence
Present governed data through notebooks and dashboards — let analysts explore ad-hoc and business users consume published metrics through a unified UI experience.
Could not build diagram for model "example-data-processing".
Notebook + SQL editor for ad-hoc analysis against the query engine.
Drag-and-drop dashboard builder backed by the semantic layer.
Both surfaces query data through Serving — exploration uses raw SQL via the Query Runner; dashboards resolve governed metrics via the Chart Renderer and Semantic Layer.
The full analyst journey — notebook → dashboard → publish:
Live tiles on published dashboards refresh from stream outputs:
Self-service with guardrails
Analysts get raw SQL in exploration mode; published dashboards always route through the semantic layer so business metrics stay consistent.